Performance of Computer Communication Systems: A Model-Based Approach

By Boudewijn R. Haverkort

Performance of Computer Communication Systems A Model-Based Approach Boudewijn R. Haverkort Rheinisch-Westfälische Technische Hochschule Aachen, Germany Computer communication systems and distributed... This description may be from another edition of this product.

• Format: Hardcover
• Language: English
• ISBN: 0471972282
• ISBN13: 9780471972280
• Release Date: December 1998
• Publisher: Wiley
• Length: 520 Pages
• Weight: 2.20 lbs.
• Dimensions: 9.7" x 1.4" x 6.8"

Customer Reviews

Probably the best modeling book around!

I consider this book a lucky find. As you can probably guess, I have nearly every book remotely related to computer performance modeling, which you'll see if you look at my guide to building a web site that doesn't frustrate customers. But for some reason, I didn't have this one. When I did get a copy, it quickly became one of my favorites! The book starts with a good foundation in stochastic processes (Part I) and single-server queuing models (Part II). The highlight of the second part is chapter 8 on phase-type distributions, quasi-birth-death processes and matrix geometric methods. I have not found much on this approach elsewhere that's accessible to a practitioner, so if you, like me, have a large collection and are looking for a reason to add this book, here's one! Part III covers queuing networks. There isn't much here you won't find in other books, although it's explained in more detail and covers more of the history of the subject than many of the basic books. Few of today's books go beyond mean value analysis for closed networks, for example; it's very rare that the convolution algorithm is even mentioned. Part IV covers stochastic Petri net models. Most of this, except for chapter 17 on infinite state stochastic Petri nets, is available in other books. But, like chapter 8, you won't find the material in chapter 17 in an accessible form anywhere else. And a number of the other books on stochastic Petri nets have gone out of print. So chapter 17 is reason two to add this one to a collection. Part V consists of one chapter on simulation. I personally think simulation has gotten way too much bad press. Analytical modeling proponents are fond of whining about how long it takes to build simulation models, how long it takes to run them, etc. I don't personally do simulation, but there are numerous situations where it's the "right thing to do". In summary, I think this is currently the best advanced modeling book available. If you want to go beyond mean value analysis, get this one. It covers stochastic processes, single queues, queuing networks, stochastic Petri nets and simulation.